Computed with the open FitLLM engine — accurate per-layer KV-cache modeling, not a naive estimate. Updated 2026-07-10.
Ranked by memory size (a proxy for cost and availability), not price. Every figure is computed by the engine — these are a floor, not a guarantee; leave headroom for your runtime and OS.
~4-bit ≈ Q4_K_M (GGUF, llama.cpp) on GPU / 4-bit (MLX) on Mac · ~8-bit ≈ Q8_0 / 8-bit · KV cache F16 · "full" = the model's max context (203K).
| Setup | Smallest GPU | Smallest Mac |
|---|---|---|
| ~4-bit · 8K ctx | ✅ RTX 3090 · 21.9/24 GB | ✅ M5 Pro 48GB · 24.4/48 GB |
| ~4-bit · full (203K) | — 🔴 | ✅ M5 Max 64GB · 41.5/64 GB |
| ~8-bit · 33K ctx | — 🔴 | ✅ M5 Max 64GB · 42.2/64 GB |
| ~8-bit · full (203K) | — 🔴 | ✅ M5 Max 128GB · 57.1/128 GB |
| Hardware | Memory | Max context (~4-bit) | Used @8K |
|---|---|---|---|
| RTX 3060 12GB | 12 GB | ❌ won't fit | 21.9 / 12 GB |
| RTX 4080 SUPER | 16 GB | ❌ won't fit | 21.9 / 16 GB |
| RTX 5080 | 16 GB | ❌ won't fit | 21.9 / 16 GB |
| RTX 3090 | 24 GB | ✅ up to 19K | 21.9 / 24 GB |
| RTX 4090 | 24 GB | ✅ up to 19K | 21.9 / 24 GB |
| RTX 5090 | 32 GB | ✅ up to 105K | 21.9 / 32 GB |
| M5 Pro 48GB | 48 GB | ✅ up to 168K | 24.4 / 48 GB |
| M5 Max 64GB | 64 GB | ✅ up to 203K | 24.4 / 64 GB |
| M5 Max 128GB | 128 GB | ✅ up to 203K | 24.4 / 128 GB |
GLM-4.7-Flash uses MLA (Multi-head Latent Attention): K/V are compressed into a single low-rank latent (512 + 64 RoPE dims) shared across all heads — cached once, not per-head K and V. Naive "2 × heads × head_dim × layers" formulas over-count its KV cache by an order of magnitude. So GLM-4.7-Flash has no single fixed memory requirement — it shifts with quantization and context. See the full breakdown.
▶ Open the interactive calculatorAll numbers are computed by the open-source fitllm-engine (MIT) from official model config.json values — reproduce or audit them yourself. Estimates; real usage varies with runtime (llama.cpp / MLX / Ollama), driver and display. Found a mismatch? Report it. · FitLLM home